33 results about "Dynamic filtering" patented technology

The application discloses a railway catenary foreign matter detection method based on frequency-space cooperation purification, and relates to the field of railway engineering technology.The method comprises the following steps: at the front end of the main network, input shallow feature maps are respectively processed by a spatial domain adaptive compensation branch and a frequency domain dynamic filtering branch, and a purified feature map is obtained through fusion; the bottleneck layer of a feature extraction module in the main network is reconstructed, and linear topological features are extracted by using a multi-scale parallel strip convolution path; a spatial attention map is generated by a shape perception substream for the feature map output by the neck network, pixel-level weighting is performed on the regression substream features, and a bounding box prediction result is output, and the network is trained by using a total loss function comprising a classification loss, a bounding box overlap loss and a shape penalty loss.The application can inhibit regular background noise, improve the continuity of foreign matter feature expression, and help improve the positioning accuracy of small foreign matters.

A method of noise reduction, including: collecting configuration data with agents from configuration items, transmitting the collected configuration data to a server for processing, blocking receipt of selected configuration data by a server load protection module to protect the server from overload, storing the configuration data at the server in a storage, processing the configuration data with a batch noise detection module that identifies noise in the configuration data; the batch noise detection module deduces rules based on the identified noise and stores the rules in a data knowledge repository, receiving a user query from a user using a front-end module, analyzing the configuration data with an analytical module to transform the configuration data to be aligned with the user query, responsive to the analysis applying a runtime noise detection module to identify and suppress transformed configuration data to remove irrelevant transformed configuration data.

The present application relates to a kind of well gas invasion and loss parameter inversion methods based on unscented Kalman filter, belong to petroleum engineering technical field, comprising: step 1: establish overflow wellbore multiphase flow model;Step 2: establish inversion model;Step 3: solve inversion model;Calculate the inversion result of downhole gas invasion and loss parameter.The application method combines wellbore multiphase flow equation with state space model, uses the nonlinear estimation characteristics of UKF algorithm, carries out dynamic filtering and state updating to wellhead measurable data, realizes the real-time inversion of downhole gas volume fraction, gas invasion flow and loss amount.UKF method can accurately capture the nonlinear characteristics of system without linearization, thereby improving the inversion accuracy and calculation stability under complex conditions.Through example verification, this method has good applicability under single reservoir, multiple reservoir and overflow and leakage coexistence conditions, provides reliable boundary conditions and intelligent control basis for well control system.

Examples provide a system and method for dynamically filtering candidate item identifiers (IDs) from a pool of item IDs in real-time for automatic labeling of images for use as training data used to train computer vision (CV) models. Images of carts are paired with item receipts. Candidate item IDs are extracted from the receipts. Item recognition inference results generated by CV models are used to pair images of individual items with item IDs identifying the item in each item image. As each candidate item ID is assigned to an item image, the item ID is dynamically filtered. Any candidate item IDs remaining after filtering are assigned to any item images failing to pair with an item ID based on the infer results. The results are presented for review and status update via a user interface device for faster and more accurate auto-labeling of training data for CV models.

This application provides a nighttime surveillance video enhancement method based on spatiotemporal joint and photometric adaptive mapping, belonging to the field of video image processing. The method includes: firstly, constructing a temporally correlated video frame sequence; then, decoupling the video sequence into a low-frequency illuminance feature sequence and a high-frequency reflectance feature sequence through a dual-branch decoder network; next, applying an iterative photometric adaptive mapping function to adjust the brightness of the low-frequency illuminance features, and applying deformable convolution and three-dimensional dynamic filtering to perform spatiotemporal alignment and denoising processing on the high-frequency reflectance features; finally, fusing the processed illuminance and reflectance features to reconstruct the enhanced video frame. This invention solves the problem of inter-frame temporal flicker and noise avalanche caused by brute-force brightening in traditional algorithms by completely decoupling and differentially parallelizing illuminance and reflectance components, improving details in dark areas while preserving edge sharpness, and achieving high-quality real-time enhancement of nighttime surveillance videos.

The application discloses a digital filtering reconstruction method for lossless audio, and relates to the field of audio filtering reconstruction. Firstly, the transient vibration intensity of the segmented audio frame is extracted in real time to obtain a transient index value, and the transient index value is mapped into a window asymmetry factor by using a nonlinear control curve. According to the factor, a front half cycle attenuation compression and a rear half cycle stretching are applied to a reference window function to generate a dynamic asymmetric window and further obtain dynamic filter coefficients. Finally, the audio is weighted and convoluted by using the dynamic coefficients, and a cross-fading splicing technology is combined to eliminate frame boundary artifacts caused by dynamic adjustment. In this way, dynamic reconstruction can be realized only by relying on time-domain point-by-point modulation of a single set of filters, and the storage of a large number of filter banks or complex offline conversion requirements are completely eliminated.

This invention provides a lightweight adaptive attack chain extraction method, system, and storage medium for cross-text-log heterogeneous data sources. The method includes: Step 1: Attack chain extraction across text-log heterogeneous data sources; automatically extracting standardized attack technique chains and tactical chains from unstructured CTI reports and structured network alarm logs; Step 2: Adaptive prediction of tactical chains based on TCKC graph constraints and multi-reward PPO; dynamically predicting the attacker's next tactic using reinforcement learning to generate a high-confidence tactical evolution path; Step 3: Attack technique chain prediction based on sequence modeling and dynamic filtering using mapping rules; guided by the tactical chains predicted in Step 2, using a Transformer model to predict the specific technique sequence the attacker will use. The beneficial effects of this invention are: improved automation capabilities for heterogeneous data sources and realization of high-confidence dynamic tactical prediction.

This invention discloses an overheating warning method for distribution boxes with integrated wireless modules, relating to the field of overheating warning technology for distribution boxes. The method includes: acquiring temperature signals and performing dynamic filtering and drift compensation processing; dividing the processed signals into regions and obtaining complete temperature field data through structurally constrained spatial interpolation to identify local hotspots; constructing a thermodynamic model for local hotspot regions and determining an adaptive temperature threshold by combining temperature safety margin and temperature rise trend correction; comparing real-time temperature data with the adaptive temperature threshold, triggering an overheating warning using a dual judgment mechanism of continuous over-threshold cumulative counting and duration, and conducting risk assessment based on temperature recovery capability. This invention solves the problems of difficulty in accurately identifying the overall temperature distribution and overheating risk inside the distribution box, and the tendency for false alarms or delayed warnings, caused by limited sensing nodes, susceptibility to temperature data interference, and fixed threshold judgments.

The application discloses a two-way ground positioning method based on three-dimensional coordinate rotation compensation and dynamic filtering, relates to the technical field of navigation positioning and space measurement and control, and comprises the following steps: collecting phase differences obtained by all ground stations during one two-way ground positioning; obtaining a spacecraft state space equation according to dynamic modeling, further obtaining a discrete form of a state updating equation according to the spacecraft state space equation, and predicting a current spacecraft predicted state according to historical information by using the state updating equation; obtaining observation values of the spacecraft, correcting the current spacecraft predicted state by using the observation values, and obtaining a current spacecraft estimated state, wherein the current spacecraft estimated state comprises a spacecraft position estimated value and a spacecraft speed estimated value; obtaining a transmission delay according to the spacecraft position estimated value, correcting a spacecraft coordinate into a geocentric geodetic coordinate system at a positioning time, and obtaining a three-dimensional position and speed of the spacecraft in the geodetic coordinate system at the positioning time after the coordinate system is corrected.

The present application belongs to the technical field of radar target simulation, and relates to a high-resolution near-range radar target simulation method and system based on a DRFM. The present application completes target modeling through a dynamic filter network, generates a target feature sequence that dynamically changes with a radar field of view; completes inherent delay calibration and cross-cycle compensation parameter calculation through measurement; after receiving a radar transmission signal, obtains a zero intermediate frequency IQ signal through frequency conversion, sampling and mixing; completes convolution modulation of the IQ signal and the target feature sequence through an adaptive segmented convolution algorithm, generates a high-resolution target echo signal; superimposes Doppler frequency shift and distance offset, completes digital-to-analog conversion and radio frequency up-conversion after cross-cycle compensation, and outputs a target echo radio frequency signal. The present application greatly reduces the calculation complexity and hardware cost of high-resolution target simulation, realizes multi-signal system adaptation for super-near-range simulation, effectively suppresses Doppler spread distortion in a large bandwidth scene, and has wide applicability.

The application discloses a low-light enhancement method and device based on a dynamic filter and a storage medium, and belongs to the technical field of image processing. The low-light enhancement method based on the dynamic filter first extracts different scale features of an input low-light image by using a residual down-sampling module and a pyramid convolution pooling module in an encoder to obtain multi-scale image feature information; then, the feature information of different scales is input into a filter generation network to obtain each local dynamic filter at a decoding end, and the image feature information obtained by the encoder is input into a decoder to perform decoding processing by using each local dynamic filter and a pixel recombination up-sampling module, so that low-light enhancement and blur removal are realized. The method has the characteristics of multi-scale feature extraction, feature enhancement, spatial change adaptability and structure reservation, and solves the problem that insufficient light greatly reduces the visual quality of an image, and further causes image detail blur, contrast reduction and noise.

This invention relates to the field of image processing technology, specifically to an image inpainting method based on wavelet-enhanced Mamba and multi-domain feature learning. In this invention, Discrete Wavelet Transform (DWT) is used to decompose image features into low-frequency and high-frequency sub-bands, achieving collaborative restoration of structure and texture in different frequency domains. An LFSC-Mamba module is designed for the low-frequency sub-band, combining Mamba's linear sequence modeling capability with a channel-space dual attention mechanism to achieve global structure reconstruction while maintaining linear computational complexity. An HFDP module is designed for the high-frequency sub-band, employing a dynamic filtering kernel generation strategy in the Fourier domain and combining it with a multi-head attention mechanism to achieve fine-grained synthesis of directional and pattern textures, significantly improving the realism and diversity of texture generation.

A dynamic filtering method for tire slip ratio signals based on Gaussian process regression (GPR) belongs to the field of tire dynamics data processing and modeling technology. The method includes the following steps: Step 1: Extracting data of the tire slip ratio and the corresponding longitudinal force dynamic region; Step 2: Calculating the rate of change of slip ratio over time and constructing an input feature vector of slip ratio and its rate of change; Step 3: Predicting the longitudinal force signal using a master Gaussian process regression (GPR) model; Step 4: Constructing and training a residual Gaussian process regression (GPR) model to compensate for the prediction error of the master model; Step 5: After performing butter filtering on the low-frequency slip ratio signal, using the trained GPR model combination to predict the low-frequency dynamic longitudinal force. This invention effectively captures the dynamic hysteresis characteristics in the tire slip ratio signal, significantly improving the dynamic accuracy of longitudinal force prediction; and through dual-model collaborative prediction, it greatly reduces the impact of high-frequency dynamic noise on the accuracy of the tire model.

This application discloses a method, apparatus, storage medium, and electronic device for processing game data. The method includes: responding to a received match merge request; determining the matching degree of candidate match instances based on the difference in match progress between the requesting match instance and candidate match instances, the dynamic intensity of the match, and the predicted probability that the candidate match instance will accept the match merge request; selecting a target match instance from the set of candidate match instances based on the matching degree and a dynamic filtering radius; performing conflict detection on the merged operation data after merging the request and target match instances based on the historical operation data of the requesting match instance and the historical operation data of the target match instance to obtain global state data; and sequentially executing each operation in the merged operation data according to the round in the global state data. This application solves the technical problem of low operability in turn-based game matches.

This invention relates to the field of voice interaction technology, and discloses a method and system for responding to dining needs based on voice interaction. The method includes: acquiring the coordinates of the user's voice source and the frequency domain features of ambient noise; constructing a dynamic filtering mask based on the frequency domain features of ambient noise; extracting semantic segments of the needs based on the user's voice source coordinates; performing structural completion on the semantic segments of the needs based on a service conflict rule base; generating a set of service elements; performing conflict detection on the set of service elements and the waiter's status; generating a service path allocation weight table; driving the interactive terminal to output service guidance instructions according to the service path allocation weight table, and simultaneously updating the dynamic load coefficient of the waiter's status; monitoring the synergy between the dynamic load coefficient and the service path allocation weight table; and triggering a secondary demand response compensation decision. This invention can solve the problems of service response delays and errors caused by unclear speech recognition, inaccurate semantic understanding, and suboptimal manpower scheduling in a restaurant environment.

This invention relates to the field of data processing technology, and in particular to a data visualization processing method applied to smart water conservancy. The method includes: collecting data and generating a hydrological dataset; processing the data to generate a rainfall raster matrix; generating a spatial credibility marker set by comparing topographic reflectance through a verification channel; detecting rainstorm conditions based on the marker set and activating an anti-interference channel; reconstructing the rainfall sequence using a particle swarm optimization algorithm; filling the spatiotemporal gaps in radar data using a generative adversarial network; controlling the core parameters of the hydraulic model; selecting an interpolation strategy based on the projection deviation value; combining the piping probability with a pulsating red sphere holographic projection to generate risk visualization parameters; isolating high-risk operations to a sandbox using a dynamic filtering mechanism; triggering disaster scenario retrieval through pulsating distribution clustering; and optimizing the topographic reflectance benchmark library based on the retrieval results and feeding it back to the verification channel. This invention effectively eliminates visualization distortion caused by spatiotemporal drift of multi-source data and blocks the contamination transmission chain of erroneous data to the model training set.

This invention discloses a method, system, platform, and storage medium for generating and processing tunnel orthophotos based on global constraint optimization and contour refinement distance determination. It generates and acquires first data corresponding to the tunnel; based on a longitudinal geometric reference with known physical spacing, it uses a global constraint optimization method to longitudinally calibrate the first data, generating corresponding second data; it then refines the contour distance determination for each section in the second data, generating corresponding third data. This method overcomes the limitations of local correction in dynamic filtering, achieving global compensation for longitudinal mileage errors and improving the geometric consistency of long-distance tunnel images. It accurately extracts the main lining contour in complex sections, effectively eliminating interference from auxiliary structures such as pipelines and cable troughs, ensuring that the distance division is based on the actual tunnel wall. Based on the distance division, it achieves adaptive grayscale equalization, compensating for intensity unevenness caused by changes in laser incident angle, point density, and curvature, thus improving image visual quality and detail recognition.

The application relates to a three-dimensional scene reconstruction method and system based on three-dimensional Gaussian splashing, wherein the method performs density self-adaptive block on a to-be-reconstructed scene through a sparse point cloud, constructs a directed bounding box for each sub-block to form a training domain of each sub-block, assigns a corresponding training view set to the training domain of each sub-block based on the visibility relationship of the sparse point cloud and performs perspective balance sampling to perform three-dimensional reconstruction on each sub-block, divides the region of all overlapping three-dimensional Gaussian primitives in the overlapping region of adjacent sub-blocks, deletes the overlapping three-dimensional Gaussian primitives in the exclusion area, dynamically filters the overlapping three-dimensional Gaussian primitives in the competition area according to the competition score result of the overlapping three-dimensional Gaussian primitives, realizes cross-block random stitching based on the overlapping three-dimensional Gaussian primitives in the competition area after dynamic filtering and the overlapping three-dimensional Gaussian primitives in the core area, and generates a three-dimensional reconstruction model of the to-be-reconstructed scene. Thus, the application realizes seamless stitching of the three-dimensional reconstruction model and guarantees continuity.

The application discloses an infrared and visible light image fusion method based on degradation perception and frequency integration, belongs to the technical field of image fusion, and is used for image fusion. The method comprises the following steps: obtaining an infrared image and a visible light image; inputting an encoder; performing fast Fourier transform; obtaining phases and amplitudes and respectively performing dynamic filtering; using a nonlinear mapping module to perform channel splicing on the dynamic filtering results of the two phases to obtain a fusion phase spectrum; and performing inverse fast Fourier transform on the fusion phase spectrum and the dynamic filtering result of the amplitude to obtain a fusion feature; inputting the fusion feature into a degradation perception expert prompt module; performing feature multiplication on the fusion feature and a scaling factor and performing feature addition on the fusion feature and a bias parameter to complete modulation of the fusion feature; and inputting the fusion feature into the encoder to obtain a fusion image. The application significantly improves the infrared and visible light image fusion quality, enhances the expression of a structure sensitive frequency band, effectively retains key texture and geometric information, and significantly improves the structure fidelity of the fusion image.

A dynamic filter for a radiographic imaging system includes a frame having an interface member structured for coupling the dynamic filter to an x-ray source of the radiographic imaging system, a movable member made of an x-ray blocking material that is structured to block a portion of an incident x-ray beam of the x-ray source to prevent it from travelling to an x-ray detector of the radiographic imaging system, and a motor supported by the frame and coupled to the movable member, wherein the motor is structured and configured to control operation of the motor to move moveable member based on data indicative of a position of and / or movement of a patient

The application provides a dynamic equalization method and circuit, which comprises the following steps: calculating a maximum gain coefficient and a minimum gain coefficient according to an input signal and preset processing parameters; multiplying the filtered input signal and one of the maximum gain coefficient and the minimum gain coefficient to obtain a first output signal; multiplying the input signal directly or after filtering or delaying and the other of the maximum gain coefficient and the minimum gain coefficient to obtain a second output signal; and superimposing the first output signal and the second output signal to obtain a final output signal; and adjusting the proportional relationship between the two input signals to achieve the purpose of adjusting the final output signal in real time. This method simplifies the required calculation amount and storage space of the dynamic filter, avoids the problem of harmonic and noise caused by the sudden change of the filter coefficient, and is effective in optimizing the calculation amount.

The present application belongs to the technical field of intelligent transportation, and particularly relates to a traffic state prediction method based on an adaptive dynamic space-time graph convolution network, which comprises the following steps: firstly, adaptively inferring a graph structure from macro and micro perspectives; secondly, extracting time dependence by using a multi-scale gated TCN; thirdly, perceiving a historical change trend hidden in a traffic state data sequence by a time trend perception self-attention mechanism, so as to realize accurate long-term prediction; and finally, generating a dynamic filter at each time step to filter node embedding and generate a dynamic graph, and capturing dynamic spatial dependence by combining an adaptive adjacency matrix. The present application can well mine complex space-time correlation in traffic state data, so as to reveal potential space-time correlation of a dynamic traffic system. Extensive experiments are conducted on two real traffic state data sets, and the experimental results show that the present application has achieved a good prediction level.

The present application relates to the technical field of vegetation maintenance management, and particularly relates to a concave green land vegetation management system based on a UAV, which collects current images of a concave green land area and extracts state information through a UAV collection module, filters analog images and constructs a feature explicit image set through a feature matching module, filters a dynamic representation of a local area and extracts contour performance information of vegetation leaves based on changes in pixel point brightness of the local area through a dynamic filtering module, calculates a correlation coefficient of the contour performance representation and the changes in pixel point brightness to determine whether vegetation state monitoring is effective through a growth monitoring module, and determines a vegetation management strategy for the dynamic representation of the local area according to water aggregation through a processing module. Thus, the accuracy of the determination of the growth state of vegetation under different water aggregation terrains of the concave green land is improved, and the scientificity of the execution of the vegetation management strategy is enhanced.

This application discloses a data processing method, apparatus, device, and storage medium, relating to the field of database technology. It is used to automatically determine whether to perform dynamic filtering on SQL query statements, helping to reduce the manual cost of performing dynamic filtering and avoiding "negative speedup." The method includes: receiving a Structured Query Language (SQL) query statement and determining the execution information of the SQL query statement; the execution information includes at least one of the following: multi-table join information of at least two tables, query condition information of the SQL query statement, and query field information of the SQL query statement; predicting a first duration and a second duration based at least on the execution information; wherein the first duration is the execution time of the SQL query statement when dynamic filtering is performed, and the second duration is the execution time of the SQL query statement when dynamic filtering is not performed; and performing dynamic filtering on the SQL query statement if the first duration is less than or equal to the second duration.

The application belongs to the technical field of general crystal material generation, and discloses a general physical law injection condition diffusion crystal generation and preference alignment method. A variational autoencoder is used to encode the full atomic structure of molecules and crystals into a shared latent space, and a diffusion model generates new latent variables thereon, which are restored into target structures by a decoder to realize unified representation modeling of heterogeneous data. A set of physical constraint alignment strategies based on direct preference optimization is constructed. Direct preference optimization is introduced into the crystal generation scene, a physical feasibility judgment index system is constructed, multi-dimensional judgment results are aggregated into sequence-level quality preference signals, and a dynamic filtering mechanism is combined to screen high-confidence mixed reward samples to calculate strategy gradients, and the diffusion generation model is fine-tuned for preference alignment, thereby significantly improving the rationality of the generated structure. The application effectively solves the structure effectiveness and physical rule alignment problem in cross-domain material generation, and significantly improves the physical rationality and structure stability of the generated material.

This invention discloses a non-contact physiological signal processing method for low-latency interference identification and dynamic filtering, comprising: acquiring human temporal signals through sensors; inputting the human temporal signals into a pre-constructed interference prediction model to obtain interference probability and corresponding interference type labels; controlling the interference prediction model to softly switch to an expanded convolutional branch or a pruned expanded convolutional branch according to a calculation depth adjustment factor, and inputting the human temporal signals into the corresponding branch to obtain enhanced human features; inputting the enhanced human features into a multi-scale dilated convolutional layer to obtain rhythm fusion features; performing a fast Fourier transform on the rhythm fusion features to obtain spectral information, and then locating the main peak of the physiological rhythm based on the energy distribution according to the spectral information to complete the decoding of physiological parameters. This invention achieves the overall effect of real-time, robust, and high-precision extraction of human temporal signals as physiological parameters under non-contact conditions.

This invention discloses a dynamic filtering method for meteorological radar data based on spatiotemporal elements, comprising: Step 1, collecting raw meteorological radar data and generating meteorological radar reflectivity grid data; Step 2, constructing a dynamic meteorological radar reflectivity dataset; Step 3, collecting aircraft surveillance data and obtaining aircraft dynamic parameters; collecting aircraft flight plan data and obtaining aircraft plan parameters; Step 4, determining the aircraft flight stage by combining the aircraft dynamic parameters and plan parameters; Step 5, setting a spatiotemporal filtering window according to the aircraft flight stage; Step 6, filtering meteorological radar reflectivity data that meets the conditions from the raw meteorological radar reflectivity grid data according to the spatiotemporal window. The filtered meteorological radar data can achieve full coverage monitoring of the area surrounding the aircraft, fully meeting the pilot's real-time monitoring needs for the surrounding meteorological environment and providing reliable support for safe flight of aircraft in hazardous weather conditions.

The present application relates to separation, more particularly to the technical field of valve filter, and disclose a Y type filter with a blowdown valve, including Y type filter body, the Y type filter body is provided with filter part, the filter part is installed with sealing part, the sealing part is installed with scale prevention part, the sealing part is provided with the control of pollution part, the sealing part is connected with the blowdown part, the filter part includes front filter port, the front filter port is inserted in the filter port of Y type filter body, the inner wall of Y type filter body is fixedly connected with clamping block, the front filter port is provided with clamping groove, dynamic rotation scale prevention is adopted, static deposition problem is completely solved, the motor drives dynamic filter to rotate, makes the impurity evenly distributed and is thrown out by centrifugal force, avoids the defect that traditional fixed filter screen unilateral is blocked, compaction is scaled, prolongs the maintenance cycle of filter screen, reduces filter screen wear and tear and replacement cost.

The application provides a traffic monitoring video compression method under millimeter wave radar cooperative positioning, and belongs to the technical field of intelligent traffic and video coding; position information of a target is obtained through a millimeter wave radar, the target position in a radar coordinate system is mapped into a pixel coordinate system of a video frame, a boundary box of an ROI region is determined, and the ROI region and a non-ROI region are distinguished; for the non-ROI region, on the one hand, a gradual mask technology is used to create a smooth transition region from the ROI to the non-ROI, and on the other hand, dynamic filtering is used to adaptively apply differentiated filtering strength to different regions of the video frame, so that seamless connection of the non-ROI and detail reservation of the ROI are realized; for the smoothed video frame, dynamic QP adjustment is performed according to a bit resource allocation strategy, and more bits are allocated to the ROI region. The application solves the problems of difficulty in dynamically extracting an ROI region and how to improve overall code rate stability while ensuring the definition of a key region.